The long term goal of the "Genomics of AML" PPG is to define the genetic and epigenetic changes responsible for the development of acute myeloid leukemia. We intend to identify somatic mutations that are responsible for the initiation and progression of disease (Projects 1, 2, and 4), and the genetic and epigenetic changes associated with relapse and chemotherapeutic resistance (Project 3). We also intend to identify mechanisms leading to increased AML susceptibility in patients with bone marrow failure syndromes (Project 5), and in patients who have received alkylator therapy (Project 6). To accomplish these aims, we have developed an infrastructure that utilizes array-based whole genome screens to identify genes for resequencing. Pathologic material and clinical data from AML patients is collected in Core A, and patient samples are banked and subjected to genomic screens in Core B. A set of 94 fully annotated "discovery set" AML samples will be assessed with array-based expression profiling, and will be tested for deletions, amplifications, and uniparental disomy with high-resolution, array-based comparative genomic hybridization and SNP genotyping. Similar studies will be performed with samples obtained at relapse, and compared with samples collected at presentation (Project 3). The output of these genomic screens will be analyzed using a variety of novel approaches (Project 2 and Core C) to identify good candidate genes for resequencing in Core D. When genetic changes that are predicted to alter gene function are identified, germline samples derived from the same patients will be resequenced to determine whether the change is somatic. The frequency of mutations will be further assessed in an additional 94 carefully annotated samples obtained from CALGB. All potentially relevant mutations will be assessed for impact on clinical outcomes and gene expression patterns (Project 2 and Core C). To discover mutations that are 'invisible'to these genomic screens, we will also sequence the entire genomes of the AML cells and skin cells from at least five individuals with FAB M1 AML (Project 1). Mutations defined by this discovery pipeline will be selected for biologic validation in the laboratories of the investigators heading Projects 2-6. Using the information gained from these studies, we intend to create molecular diagnostic tools for disease stratification, and we will identify candidate genes for targeted therapeutic approaches.